Rehabilitation after an exacerbation of chronic respiratory diseaseBMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g4370 (Published 08 July 2014) Cite this as: BMJ 2014;349:g4370
- William D-C Man, consultant chest physician1,
- Samantha S C Kon, MRC clinical research fellow1,
- Matthew Maddocks, lecturer in health services research2
- 1NIHR Respiratory Biomedical Research Unit, Royal Brompton and Harefield NHS Foundation Trust and Imperial College, London
- 2King’s College London, Cicely Saunders Institute, London
- Correspondence to: W D-C Man, Department of Respiratory Medicine, Harefield Hospital, Middlesex UB9 6JH, UK
Emergency hospital admissions for exacerbations of chronic respiratory disease—principally chronic obstructive pulmonary disease (COPD)—represent major life events for patients and place an enormous financial burden on healthcare systems. Readmission after hospital discharge is common and increasingly carries a financial implication for acute hospitals in both the United States and the United Kingdom.1
Pulmonary rehabilitation is an evidence based multidisciplinary intervention comprising exercise training, education, and behaviour change that is the cornerstone of non-drug management of adults with stable COPD. A Cochrane review summarising nine small, heterogeneous trials found clinically significant improvements in physical functioning, quality of life, and readmission rates after pulmonary rehabilitation delivered during or soon after a hospital admission.2 Consequently, recent guidelines recommend this approach.3 4
In a linked paper (doi:10.1136/bmj.g4315), Greening and colleagues report the results of a pragmatic two arm randomised controlled trial comparing an early rehabilitation strategy with usual care during and after admission to hospital for an exacerbation of chronic respiratory disease.5 This trial was adequately powered, with good patient uptake (only a quarter of those eligible declined participation), intention to treat analysis with robust handling of missing data, and multiple assessments over 12 months, providing a comprehensive dataset.
Although the results were found to be largely negative, the paper is thought provoking, not least because the observations from this trial contrast starkly with those from previous trials and the current recommendations supporting pulmonary rehabilitation in the acute setting. The intervention had no effect on hospital readmissions at 12 months—the primary outcome measure. More worryingly, the intervention seemed to be associated with harm—namely, a higher risk of death.
How can these findings be explained? Unlike previous UK trials where rehabilitation was supervised and provided in the outpatient setting after hospital discharge,6 Greening and colleagues’ intervention began with an initial supervised inpatient component, followed by a remotely supervised home exercise programme after discharge. It is logical to assume that the inpatient component might help prevent the deleterious consequences of a hospital admission, such as rapid decline in daily physical activity and muscle function.7 8 However, the median length of hospital stay was only five days (which is typical of the United Kingdom), and the amount of supervised exercise during admission was modest. In addition, participants in the intervention group reported poor adherence to daily training during the home component, and poor uptake of the formal outpatient based pulmonary rehabilitation offered to both groups after three months (14% compared with 22% in the group that received usual care). This health behaviour is interesting as it suggests that patients receiving the early intervention may have considered their rehabilitation needs to have already been met by the service.
Both groups had comparable changes in physical functioning over 12 months, which underscores the limited exercise training delivered within the intervention. This may partly explain the lack of effect on hospital readmissions. Evidence is emerging of a clear link between poor physical function, inactivity, and increased risk of admission to hospital among adults with COPD.9
The extra deaths observed in the intervention arm are harder to explain. Pulmonary rehabilitation is generally considered safe, and reports of serious adverse events in clinical practice are rare. The Cochrane review of rehabilitation after acute admission for exacerbations of COPD suggested that pulmonary rehabilitation may in fact reduce mortality.2
It seems unlikely that the higher mortality at 12 months reported by Greening and colleagues was due directly to an intervention that had an exercise component of such modest intensity and duration. The most likely explanation is the case mix; the intervention and control groups were probably not matched at baseline for comorbidity or frailty—a state that increases the vulnerability to adverse outcomes but is notoriously difficult to quantify reliably.10 None the less, the possibility that the intervention subtly influenced health behaviour to the detriment of participants should still be considered. A recent trial of a largely unsupervised, self management intervention in patients with COPD was stopped prematurely because of an unexpected excess mortality associated with the intervention.11
The trial by Greening and colleagues tackles a major healthcare problem and provides valuable insights. Firstly, given the increasingly short duration of inpatient stays for exacerbations of chronic respiratory disease in the United Kingdom, an inpatient exercise rehabilitation strategy is unlikely to produce major health gains and confers no additional benefit over standard physiotherapy and usual care alone. However, this should not be extrapolated to other healthcare systems where inpatient rehabilitation is longer and better accepted by healthcare planners. Furthermore, inpatients with chronic respiratory disease may be a captive audience and receptive to lifestyle changes, such as smoking cessation, nutrition, and increasing daily physical activity. Secondly, home based, lightly supervised rehabilitation after discharge also seems to have limited benefits, and may even cause harm.
We are therefore directed back to more intensely supervised pulmonary rehabilitation programmes, which are clinically effective3 but might not be acceptable to some patients or healthcare referrers, with an uptake of less than 10% observed in real life settings.12 Research to help identify and overcome the barriers to improving current clinical services is paramount.
Finally, this trial reveals substantial gaps in our knowledge about how best to provide exercise training for adults admitted to hospital with chronic respiratory disease. We have yet to establish the optimal modality, setting, timing, or duration of programmes, or to identify the subgroups most likely to benefit and least likely to be harmed. These should be our research priorities for the future.
Cite this as: BMJ 2014;349:g4370
Competing interests: We have read and understood the BMJ Group policy on declaration of interests and declare the following interests: none.
Provenance and peer review: Commissioned; not externally peer reviewed.